Skip to main content
trading shots

Traders work on the floor of the New York Stock Exchange on Aug. 19, 2013.BRENDAN MCDERMID/Reuters

Stocks are seriously overvalued according to the cyclically-adjusted price-earnings (CAPE) ratio developed by Yale University professor Robert Shiller. This measure stands at 23.6 compared to its long-term average of 16.5, implying U.S. stocks need to fall 40 per cent in order to revert to the mean.

But I'm not overly concerned.

CAPE compares the prices of S&P 500 stocks to 10-year moving averages of earnings in order to smooth out the impact of the business cycle. But as with all valuation yardsticks, it cannot say when overvaluation will be corrected. In the past, the catalyst has usually come when U.S. monetary and fiscal policy shifts from a highly stimulative stance (like now) to a highly restrictive one (aimed at cooling off an overheating economy). This transition usually takes 18 to 30 months, during which time bond yields and stock prices typically trend upward.

Even if an investor doesn't reduce their exposure to stocks before the bear market eventually arrives, CAPE suggests they should still be okay as long as they have a long-term horizon. Since 1880, whenever the measure has been as high as it is now, investors have enjoyed average annual real rates of return on stocks of nearly 3 per cent over the ensuing decade or two. That's below the long-run average return of 6.5 per cent, but still respectable.

But then there are a host of measurement issues. It's hard to get apples-to-apples comparisons of price-earnings ratios over a period spanning 1880 to the present. Distortions are created by changes in underlying parameters such as market structure, inflation, interest rates, taxes and accounting rules.

For example, the CAPE measure may additionally smooth out secular improvements in earnings and generate misleading signals. Of note, say analysts, globalization has trimmed the power of labour unions and led to a sustained expansion in profit margins. Such earnings growth can have a longer shelf life than cyclically inflated earnings, so by diluting its impact, CAPE might make stocks look more expensive than warranted.

Second, inflation and interest rates are currently at very low levels. Most analysts acknowledge that such an environment supports higher price-earnings ratio.

Third, corporate income tax rates in the U.S. have climbed from zero (before 1911) to 40 per cent as the welfare state was rolled out. This lowers earnings in recent decades relative to earlier decades. One analyst recalculated CAPE with a constant tax rate and found overvaluation was less alarming.

Fourth, Professor Jeremy Siegel, author of Stocks for the Long Run, argues that accounting changes made in the 1990s created a significant distortion by greatly exacerbating profit declines during the recessions in 2002 and 2008. Without these accounting changes, the 10-year average for earnings in the CAPE denominator would be a lot higher (and the CAPE ratio substantially lower). In fact, if the profit series from the U.S. National Accounts is used instead, there would be little overvaluation.

In sum, CAPE might be useful for long-term horizons, with the caveat that its projection of real returns is probabilistic, not certain. Hence, at CAPE's currently high readings (implying only average annual real returns of 3 per cent over the next decade or so), some buy-and-hold investors may decide to have less stock exposure in their strategic asset allocations. Nevertheless, for many others, CAPE has a number of measurement issues that raise doubts about its reliability as a guide to the decades ahead.

When it comes to foreshadowing turning points in the stock market, however, there are even greater grounds for skepticism. A number of other factors besides valuation – particularly central banks' monetary policies – also play crucial roles in determining market direction.

READERS: Are stocks too hot for your taste? If so, how are you limiting exposure?